Danny Sullivan Dares Google and Yahoo to Grow Up.

Source Title:
Screw Size! I Dare Google & Yahoo To Report On Relevancy
Story Text:

That's it! - Even the ultimate laid back man of SEO, Danny Sullivan, has decided the pigtail pulling has gotten boring

This is Search Engine Size Wars VI, by my count. It's absurd. It's annoying. It's a friggin' waste of time. Instead of advancing to a commonly accepted relevancy figure, the search engines want to keep us mired in the mud of who's biggest.

This game is going to go on and on until someone is brave enough to change the rules. I'm daring either of the leaders, Google or Yahoo, to do just that. Both of them say that size is one of only many factors to consider. Both of them tell you relevancy matters most. SO PROVE IT!

The problem is - knowing what matters most and delivering it are two very different things aren't they?


Size Does Matter

Now they're fighting over who's the biggest. What are we in high school?

I totally agree with Danny Sullivan. Who cares about size. And also, everytime Google (or Yahoo) announces something new, the other counters with the same offering or announcement. I'm tired of it.

Let's get back to relevancy and providing better search results.

I agree type0, i've been

I agree type0, i've been saying forever that this tit for tat bollocks should end - im (obviously) 100% with Danny also.

So is Tim Bray it would seem

it’s pitiful because right now, Yahoo and MSN are showing signs of giving Google the first serious run for its money in recent memory. So those guys should stay away from these juvenile distractions.

Perfect Example

Google's Chris DiBona Slashdotting an NCSA Study that makes Yahoo's claims look suspect.

Suspect 'third party' study

Interesting comments on the NCSA study by Sethf:

Much as I hate to admit it....

Tim has a point. That NCSA study is rubbish. The methodolgy is utter rubbish. I can see what they were trying to dobut they have demonstrated a lamentable lack of understanding of how the Web works, and the challenges faced by a modern crawling search engine.

I have some scientific training, and I can tell you that their "research" is fundamentally flawed, because it does not seek to examine other possible causes or explanations for the data. They have made some assumptions that need to hold true for the data, and more importantly, their interpretation of it, to have any value, and that is a crime against the scientific method.

Seths got a point there, but has missed, IMO, the biggest flaw. Lets have a quick peek at the NCSA paper :

the standard method to measure relative size was developed by Krishna Bharat and Andrei Broder in 1998

Right. So you are using 7 year old ideas to measure todays Yahoo? That's the rough equivalent of using Newtons work to check results from a super-collider. Pointless. We've moved on

instead of focusing on documents that match the common "web words", we chose to focus on the more obscure documents of the web – the “long tail” of the search index

Sounds good, doesn't it? Can you see the problem? Hmmmm, what if Google and Yahoo have a different view of the Web? What if Yahoo have cracked some of the deep web problem. After all, it has long been thought that SEs' were only indexing the tiniest fraction of all available information. Why should Yahoo not simply have worked out how to get a bit deeper?

If that were the case, the "long tail" assumption is worthless - the NCSA are simply measuring the wrong thing to get a reliable answer. My objection is that they seem not even to realise that this *could* be a problem, and have therefore made no effort to address it, let alone looked for, or even suggested ways to get around it. From a purely scientific point of view then, this rather limits the value of the research.

Whilst I realise that the 1000 result limit presents a problem, the chosen method of getting around it effectively destroys the value of the exercise, since it ignores a potential (and in my view, fairly likely) explanation


Of course it's stupid.

But so's whining about it.

No-one believes the press releases; no-one cares. Move on ...

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.